Total space PLS transmission method and system for STAR-RIS auxiliary wireless communication system

Abstract

The invention provides a full-space PLS transmission method and system for an STAR-RIS auxiliary wireless communication system. In the STAR-RIS auxiliary wireless communication system, a reflection coefficient and a transmission coefficient of an STAR-RIS and precoding vectors allocated by a base station to a safe user and a public user are obtained by solving a non-convex optimization problem. Wherein the non-convex optimization problem takes the maximization of the weight sum of the safety rate of a safety user and the reachable rate of a public user as an optimization target and takes the power of a base station as a constraint condition; the security rate of the secure user is obtained by performing large-scale system analysis on the average security rate of the secure user. According to the scheme provided by the invention, under the assistance of STAR-RIS, robust full-space physical layer secure transmission can be realized.

Description

Technical Field

[0001] The present invention relates to the field of new generation wireless communication technology, and in particular to a robust full-space physical layer security transmission method and system for a STAR-RIS assisted wireless communication system. Background Art

[0002] With the advent of the 5G era, driven by advanced communication and data processing technologies, people are increasingly dependent on wireless communications. A large amount of important sensitive information, such as identity information, confidential documents, etc., is transmitted in open wireless networks, increasing the risk of eavesdropping. Therefore, information security issues are receiving more and more attention. Physical layer security (PLS) based on the information theory framework is a promising solution that uses the inherent unpredictability of physical media, especially the random changes in legitimate and potential eavesdropping channels, to protect the security of wireless information transmission.

[0003] Although the existing PLS transmission schemes can effectively improve the security performance between base stations and users, these schemes are designed based on uncontrollable random channels, and the performance gains they can bring are severely limited. In order to overcome this challenge, a PLS scheme based on reconfigurable smart surface (RIS) is proposed, because RIS can dynamically adjust the electromagnetic characteristics of the incident signal to build an end-to-end controllable virtual channel between the base station and the user, greatly improving the security performance.

[0004] However, in the existing PLS scheme based on traditional RIS-assisted, RIS can only implement reflection modulation on the incident signal, which means that the transceiver terminal equipment must be located on the same side as the RIS. This will greatly limit the coverage of the wireless network and the flexibility of deploying RIS. In order to solve the above obstacles, a more advanced RIS, called simultaneous transmission and reflection RIS (STAR-RIS), was proposed, because STAR-RIS can divide the incident signal into two parts, one part of the incident signal is reflected back to the same side, and the other part is transmitted to the opposite side. Therefore, STAR-RIS has shown great application potential in wireless networks. At present, some PLS schemes based on STAR-RIS-assisted communication systems have been proposed. However, these PLS schemes generally have the following shortcomings: (1) These PLS schemes usually assume that the activity area (R zone or T zone) of the eavesdropper in the STAR-RIS service area has been predetermined, which enables the system to effectively solve the eavesdropping risk by designing appropriate security measures. It is worth noting that this assumption is unrealistic, because in practice, intelligent eavesdroppers can make full use of the full spatial coverage characteristics of STAR-RIS to enhance their ability to intercept confidential information. Specifically, they can move freely in the R area and the T area to increase the probability of successful eavesdropping, which poses a severe challenge to the current PLS schemes in maintaining information security. (2) These existing PLS schemes can only guarantee the security of semi-space wireless communications because they consider fixed eavesdroppers in a specific area and mainly guarantee the security in the R area or the T area. Note that the full-space coverage of STAR-RIS makes the STAR-RIS-assisted communication system encounter more serious inherent security issues than the traditional RIS-assisted communication system because the eavesdropper can conduct 360° monitoring of the communication between the base station and the legitimate user. Summary of the invention

[0005] In order to solve the above problems in the prior art, the present invention provides a full-space PLS transmission method and system for a STAR-RIS assisted wireless communication system.

[0006] The present invention is achieved through the following technical solutions:

[0007] A full-space PLS transmission method for a STAR-RIS assisted wireless communication system, in which a reflection coefficient and a transmission coefficient of a STAR-RIS and a precoding vector allocated by a base station to a security user and a public user in the STAR-RIS assisted wireless communication system are obtained by solving a non-convex optimization problem;

[0008] Among them, the non-convex optimization problem takes the weighted sum of the security rate of the security user and the achievable rate of the public user as the optimization goal, and the base station power as the constraint condition; the security rate of the security user is obtained by performing a large-system analysis on the average security rate of the security user.

[0009] Preferably, the non-convex optimization problem is as follows:

[0010]

[0011] Where ω1,ω2∈(0,1] represent the weighting factors of secure users and public users respectively; P tmax Indicates the maximum transmit power at the base station; λ represents the number of quantization bits; and Respectively represent the precoding vectors allocated by the base station to the security user and the public user; U r and U t denote the reflection coefficient and transmission coefficient of STAR-RIS respectively; and denote the magnitudes of the STAR-RIS reflection and transmission coefficients, respectively, satisfying denote the phases of the STAR-RIS reflection coefficient and transmission coefficient respectively; j is an imaginary unit; R c represents the achievable communication rate for public users; Indicates the security rate of the security user.

[0012] Furthermore, the safe rate of safe users for:

[0013]

[0014] Among them, H BR Indicates the channel between the base station and STAR-RIS; represents the noise power at the eavesdropping user; R b Indicates the achievable communication rate of secure users; l re represents the large-scale path loss experienced by the signal from STAR-RIS to the eavesdropping user; P1 represents the probability that the eavesdropping user is located in the R region of STAR-RIS; P0 represents the probability that the eavesdropping user is located in the T region of STAR-RIS, P0=1-P1.

[0015] Furthermore, the achievable communication rate R of a secure user is b and the achievable communication rate R of public users c They are as follows:

[0016]

[0017] Among them, h rb and h rc The channels between STAR-RIS and security users and public users are shown respectively. and represent the noise power at the safety user and the public user respectively.

[0018] Furthermore, the method for solving the non-convex optimization problem is:

[0019] Decomposing the non-convex optimization problem into an active beamforming subproblem and a passive beamforming subproblem, and solving the active beamforming subproblem and the passive beamforming subproblem respectively;

[0020] The active beamforming subproblem is expressed as:

[0021]

[0022] The passive beamforming subproblem is expressed as:

[0023]

[0024] Furthermore, the method for solving the active beamforming subproblem is:

[0025] The minimum mean square error based iterative algorithm is used to solve the active beamforming subproblem, where the active beamforming subproblem is expressed as the following convex optimization problem in the (t+1)th iteration:

[0026]

[0027]

[0028] and is the auxiliary variable in the tth iteration;

[0029] The convex optimization problem is solved using the Lagrange multiplier method to obtain the optimal solution.

[0030] Furthermore, the method for solving the passive beamforming subproblem is: using a cross-entropy optimized iterative algorithm to solve the passive beamforming subproblem.

[0031] Furthermore, the joint sampling distribution of the cross entropy optimization framework is constructed as follows:

[0032]

[0033]

[0034] in, represents the sampled solution, represents the skew parameter of the joint sampling distribution; the matrix P pq Represents φ p Select Collection The qth item in the The probability of {·} An indicator function representing an event, represents the pth entry of vector a; and denote the standard deviation and mean vector, σ m and μ m is the parameter that generates the amplitude of the mth unit of STAR-RIS; and denote the magnitudes of the STAR-RIS reflection and transmission coefficients, respectively, satisfying represents a discrete phase shift variable; denote the phases of the STAR-RIS reflection coefficient and transmission coefficient, respectively.

[0035] Furthermore, based on the joint sampling distribution, the cross entropy minimization optimization problem is:

[0036]

[0037] The Lagrange multiplier method is used to solve the above cross entropy minimization optimization problem and obtain the optimal solution.

[0038] A STAR-RIS assisted wireless communication system, in which the reflection coefficient and transmission coefficient of STAR-RIS and the precoding vectors allocated by the base station to the security user and the public user are obtained by solving a non-convex optimization problem;

[0039] Among them, the non-convex optimization problem takes the weighted sum of the security rate of the security user and the achievable rate of the public user as the optimization goal, and the base station power as the constraint condition; the security rate of the security user is obtained by performing a large-system analysis on the average security rate of the security user.

[0040] Compared with the prior art, the present invention has the following beneficial effects:

[0041] Aiming at the full-space eavesdropping challenge faced by the wireless communication system supported by STAR-RIS, the present invention proposes a robust full-space physical layer security transmission scheme. Different from the existing PLS scheme, the PLS scheme of the present invention takes into account the statistical channel information state of the eavesdropper in the base station and the uncertainty of the eavesdropper's position in the STAR-RIS assisted service area, and uses the large system analysis method to derive the asymptotic expression of the average security rate at the security user as the security rate of the security user, and formulates an optimization problem to maximize the weighted sum of the security rate of the security user and the achievable rate of the public user. The present invention can be applied to the field of reconfigurable intelligent surface assisted communication technology. Compared with the traditional physical layer security scheme, the scheme proposed in the present invention can achieve robust full-space physical layer security transmission with the assistance of STAR-RIS. Simulation results show that the proposed robust PLS scheme effectively reduces the information leakage in the coverage area of ​​the STAR-RIS system. Compared with the benchmark scheme including the assistance of the traditional reconfigurable intelligent surface, it has higher performance gain, thereby verifying the effectiveness of the robust PLS scheme and iterative algorithm.

[0042] Furthermore, by optimizing the joint design of active and passive beamforming variables under the constraints of power budget, equality constraints and actual implementation of STAR-RIS, in order to effectively solve the non-convex optimization problem with discrete variables, an iterative algorithm based on minimum mean square error (MMSE) and cross entropy optimization (CEO) is proposed. The proposed algorithm can be applied to the field of discrete optimization problem solving. At the same time, the algorithm proposed in the present invention has low complexity and excellent performance. It is particularly good at solving discrete optimization problems and has high engineering application value. The above advantages are helpful to cope with the challenges of high capacity and high security faced by future 6G networks, and bring significant economic benefits to the performance improvement of wireless communication systems. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 A schematic diagram of a STAR-RIS assisted wireless communication system of the present invention;

[0045] Figure 2 For the embodiment of the present invention, P1=P0=0.5, ω1=ω2=0.5, P tmax =30dBm,λ=2 and different N tSchematic diagram of the performance of the number of units, weights and rate changes in RIS configuration;

[0046] Figure 3 This is how the solution proposed in the present invention suppresses the leakage of STAR-RIS service area information. DETAILED DESCRIPTION

[0047] The following describes the embodiments of the present invention through specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention.

[0048] It should be noted that the process equipment or devices not specifically specified in the following embodiments are all conventional equipment or devices in the art.

[0049] It should be noted that the terms "include" and "have" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices. Moreover, unless otherwise specified, the numbering of each method step is only a convenient tool for identifying each method step, and is not intended to limit the order of arrangement of each method step or to define the scope of the present invention. Changes or adjustments in their relative relationships should also be regarded as the scope of the present invention without substantially changing the technical content.

[0050] The present invention proposes a robust full-space PLS transmission method for improving the security of a STAR-RIS assisted wireless communication system.

[0051] like Figure 1 As shown, a STAR-RIS assisted wireless communication system is constructed, including a t Antennas, a STAR-RIS with M cells, and three different users: secure user Bob, public user Carol, and eavesdropping user Eve; assume that secure user Bob is located in the R region of STAR-RIS, while public user Carol uses the communication service in the T region. Physical obstacles such as buildings and trees prevent direct contact between the base station and users;

[0052] Based on the established STAR-RIS assisted wireless communication system, the asymptotic security rate of the secure user Bob is derived, and a non-convex optimization problem is constructed with the weighted sum of the asymptotic security rate of the secure user Bob and the achievable rate of the public user Carol as the optimization objective under the constraint of base station power.

[0053] The established non-convex optimization problem is decomposed into the first non-convex sub-problem and the second non-convex sub-problem, and an iterative algorithm based on the minimum mean square error and cross entropy optimization framework is designed to effectively solve the first non-convex sub-problem and the second non-convex sub-problem to obtain the optimization result.

[0054] In the STAR-RIS assisted wireless communication system, the signal model of each user includes:

[0055] The signals received by the secure user Bob and the public user Carol can be expressed as:

[0056]

[0057] Among them, h rb and h rc are the channels between STAR-RIS and the secure user Bob and the public user Carol, respectively. BR Indicates the channel between the base station and STAR-RIS. b and c The precoding vectors assigned by the base station to the secure user Bob and the public user Carol are respectively and and are the additive white Gaussian noise at the secure user Bob and the public user Carol, and are the noise powers at the secure user Bob and the public user Carol respectively. r and U t are the reflection coefficient and transmission coefficient of STAR-RIS, which can be expressed as:

[0058]

[0059] in, and denote the magnitudes of the STAR-RIS reflection and transmission coefficients, respectively, and they satisfy denote the phases of the STAR-RIS reflection coefficient and transmission coefficient, respectively. j is an imaginary unit.

[0060] When the eavesdropping user Eve moves in the service area of ​​the STAR-RIS assisted wireless communication system, it can eavesdrop on signals in different areas. That is to say, the eavesdropping user Eve may perform eavesdropping activities in the R area, or may perform eavesdropping tasks in the T area. In order to represent the location area of ​​the eavesdropping user Eve, a Bernoulli variable b is defined. Specifically, when b=1 and the probability is P1, it means that the eavesdropping user Eve is located in the R area of ​​STAR-RIS. Conversely, when b=0 and the probability P0=1-P1, it means that the eavesdropping user Eve is located in the T area of ​​STAR-RIS. Therefore, the signal received by the eavesdropping user Eve can be expressed as:

[0061]

[0062] in, represents the additive Gaussian white noise at the eavesdropping user Eve, represents the noise power at the eavesdropping user Eve, h re Indicates the channel between Eve and STAR-RIS.

[0063] The achievable communication rates of secure user Bob and public user Carol can be expressed as:

[0064]

[0065] The eavesdropping rate of Eve on the communication between the base station and the secure user Bob can be expressed as:

[0066]

[0067] Therefore, the security rate of secure user Bob can be expressed as:

[0068]

[0069] Among them, operation [x] + Represents max{0,x}.

[0070] Considering that the eavesdropping user Eve remains hidden and moves freely within the entire coverage of the STAR-RIS assisted wireless communication system, it is challenging for the base station to obtain accurate information about the location area of ​​the eavesdropping user Eve and perfect channel state information (CSI). Therefore, the present invention chooses to use the secure user Bob at b and h re The average security rate on represents the security performance of the secure user Bob in the STAR-RIS assisted wireless communication system, which is given by the following formula:

[0071]

[0072] in,

[0073]

[0074] In fact, due to the addition of random variables h in the eavesdropping rate expression re , leading to the derivation The exact mathematical formula of is difficult to handle. To solve this problem, the present invention uses large system analysis technology, which is a common method for evaluating the performance limit of wireless communication systems. Specifically, assuming that STAR-RIS is equipped with a large number of passive units, it can be derived that:

[0075]

[0076] Among them, (a) is obtained using large system analysis techniques. It is worth noting that by performing such operations, In h re Therefore, the asymptotic expression of the average safety rate can be obtained, that is,

[0077]

[0078] Therefore, the present invention will use The asymptotic result of To represent the secure rate of secure user Bob.

[0079] The optimization problem is as follows:

[0080]

[0081] Among them, ω1, ω2∈(0,1] are the weighting factors of security users and public users respectively; P tmax Indicates the maximum transmit power at the base station; λ represents the number of quantization bits. In fact, the objective function of the above optimization problem is non-convex. In addition, it contains equality constraints related to amplitude, and there are discrete variables. There is obvious coupling between the variables, which makes the problem difficult to solve. In order to solve this problem, the original optimization problem is firstly decomposed into two sub-problems, namely the active beamforming sub-problem and the passive beamforming sub-problem, using the alternating strategy. On this basis, an iterative algorithm based on minimum mean square error (MMSE) and cross entropy optimization (CEO) is proposed to effectively solve the above problem.

[0082] The active beamforming subproblem can be expressed as:

[0083]

[0084] Note that the above active beamforming subproblem is a non-convex optimization problem, since the objective function is about (wrt)wb and w c To solve this problem, the present invention firstly converts Converted to the following form:

[0085]

[0086] Then use the MMSE method to equivalently transform f1, f2, f3, f4 and f5 into

[0087]

[0088] in,

[0089]

[0090] Here, W1, W2, W3, W4, W5, u1, u2 and u3 are introduced auxiliary variables; l re represents the large-scale path loss experienced by the signal from STAR-RIS to the eavesdropping user Eve. In fact, these auxiliary variables can be easily solved by setting the first-order partial derivative of the objective function with respect to these variables equal to zero. The optimal solution can be given as:

[0091]

[0092] According to the above analysis, the present invention can be and In the case of , the concave lower bounds of f1, f2, f3, f4 and f5 in the (t+1)th iteration are obtained, which can be expressed as:

[0093]

[0094] It is worth noting that and is the effective beamforming solution obtained at the tth iteration and Therefore, in the (t+1)th iteration, The concave lower bound of can be expressed as:

[0095]

[0096] Similarly, in the (t+1)th iteration, R c can be equivalently converted to

[0097]

[0098] in,

[0099]

[0100] Therefore, the above active beamforming subproblem is reformulated as the following convex optimization problem in the (t+1)th iteration,

[0101]

[0102] The present invention further uses the Lagrange multiplier method to solve the above convex optimization problem. Specifically, the Lagrange function of the above problem can be expressed as:

[0103]

[0104] in, represents the Lagrange multiplier (dual variable) associated with the constraint. b and w c The optimal solution of the Lagrangian problem can be expressed as:

[0105]

[0106] Among them, the operator Represents a pseudo-inverse operation.

[0107]

[0108]

[0109] The optimal solution given above shows that the optimal active beamforming and dual variables For any given The present invention can directly calculate the optimal active beamforming closed-form solution. Therefore, to obtain the optimal active beamforming of the original problem, it is crucial to solve its dual problem and obtain the optimal dual variable, because this is the key step to achieve the optimal active beamforming of the original problem.

[0110] First, the dual problem of the active beamforming subproblem is given:

[0111]

[0112] in, represents the dual function.

[0113] Regarding the dual problem, the present invention will use the subgradient algorithm to solve it. In the (l+1)th inner loop iteration, the dual variable The update rule can be expressed as:

[0114]

[0115] in, Represents the dual variable update algorithm Positive increment of .

[0116] The passive beamforming subproblem can be expressed as:

[0117]

[0118] In practice, solving the above passive beamforming subproblem poses a challenge due to the non-convexity of the objective function, the equality constraints on the STAR-RIS amplitude, and the discrete phase shifts involved. To address this problem, the CEO method is used to design the passive beamforming variables. Unlike traditional convex optimization algorithms, the CEO method uses probability-based learning techniques to solve complex problems. The basic idea of ​​​​the CEO method for dealing with complex optimization problems can be summarized as follows: (1) Design appropriate sampling distributions characterized by tilt parameters for different types of variables; (2) Generate expected solutions from the sampling distribution and evaluate these expected solutions by calculating the relevant objective function values; (3) Adjust the tilt parameters of the sampling distribution by optimizing the cross entropy between the existing sampling distribution and the benchmark sampling distribution established by these significantly effective expected solutions; (4) Iterate new feasible samples with the modified tilt parameters until the difference in the objective function values ​​between consecutive iterations drops to a predefined threshold.

[0119] The sampling distribution is designed as follows:

[0120] In order to effectively use the cross entropy optimization framework to solve the optimization problem, two appropriate sampling distribution models must first be established to generate discrete phase shift variables. and continuous amplitude variables φ p The sampling distribution model of is designed as follows:

[0121] To generate feasible phase shifts, the present invention can utilize the obtained or predetermined discrete probability distribution {P p1 ,…,P pq ,…,P pQ}In parallel and independently for each φ p , Sampling is performed, where P pq Represents φ p Select Collection The qth item in the Therefore, the sampling distribution for generating φ can be given by:

[0122]

[0123] In the formula, the matrix An indicator function representing an event, represents the p-th entry of vector a.

[0124] Continuous amplitude variable sampling distribution model: Unlike discrete phase shift, the amplitude variable is a continuous variable. Here, Gaussian distribution is the best choice for generating the amplitude variable. Therefore, the sampling distribution for generating random amplitude can be expressed as

[0125]

[0126] in, and denote the standard deviation and mean vector, σ m and μ m is the key parameter for generating the amplitude of the mth unit. Also, In practice, the magnitude of the transmission coefficient can be determined using the constraint Direct calculation, that is The sampling distribution may generate amplitudes outside the range of (0,1]. To solve this problem, the present invention can use the following techniques to ensure that the generated amplitudes are feasible: (1) remove infeasible amplitudes and randomly select amplitudes within the required range for replacement; (2) iteratively generate replacement values ​​for each infeasible amplitude until a feasible amplitude is obtained. and μ represent the variance and mean, respectively. It should be emphasized that the randomly generated f0 may exceed [f min ,f max ] range.

[0127] Joint sampling distribution model: Based on the above analysis, the joint sampling distribution for processing optimization problems can be expressed as:

[0128]

[0129] In this case, the sampling solution is sampled from the joint distribution generated in represents the skew parameter of the joint sampling distribution.

[0130] The tilt parameter update criteria are as follows:

[0131] As mentioned above, the CEO framework updates the tilt parameter by minimizing the cross entropy problem between the current sampling distribution and the benchmark sampling distribution derived from the best performing expected solution. Specifically, the present invention first generates K feasible samples based on the joint sampling distribution Then calculate the objective function value R(Ξ k ). In order to identify these efficient solutions, the candidate solutions generated They are sorted in descending order according to their objective function values. The candidate solution with the kth largest objective function value is denoted as [k] , satisfying R(Ξ [1] )≥R(Ξ [2] )≥…≥R(Ξ[k] )≥…≥R(Ξ [K] ). Before selection Samples form an elite set in Indicates the percentage of the best performing samples in the selected elite set.

[0132] The tilt parameter is adjusted by minimizing the cross entropy between the current sampling distribution and the new sampling distribution, which are derived from the elite samples in the CEO framework. The cross entropy minimization optimization problem can be equivalently transformed into:

[0133]

[0134] In fact, the Lagrange multiplier method can effectively handle the above optimization problem. In order to use this method, the present invention first constructs the Lagrangian function of the problem, which is expressed as:

[0135]

[0136] Among them, ∈ p is the Lagrange multiplier associated with the equality constraints given in the above optimization problem. The Lagrange function is P, The first-order partial derivatives of and μ can be derived as:

[0137]

[0138] let and It can be introduced:

[0139]

[0140] In order to reduce the risk of convergence to a local optimum in the CEO algorithm, a smoothing method commonly used in reinforcement learning is usually used to adjust the tilt parameter in successive iterations. The tilt parameter of the i-th iteration is modified according to the following method:

[0141]

[0142] μ (i) ←ξμ (i) +(1-ξ)μ (i-1) ,

[0143] Where ξ∈(0,1) represents the smoothing parameter. Under the adjusted tilt parameter, K new feasible candidate solutions are generated, and then the elite solution is selected to prepare for updating the tilt parameter in the next round of operation. The proposed algorithm based on the CEO framework converges to the optimal solution in consecutive iterations once the difference in the optimal objective function value between consecutive iterations is lower than a predetermined threshold.

[0144] Example 1

[0145] This embodiment examines P1=P0=0.5,ω1=ω2=0.5,P tmax =30dBm,λ=2 and different N t Considering the change in the number of antennas, the impact of the number of units deployed on STAR-RIS on the weighted sum rate is as follows: Figure 2 In particular, the present invention observes that as M increases, all cases show an upward trend, while the growth rate experiences a gradual decline. This is because more units can provide more degrees of freedom to reconfigure the wireless propagation environment.

[0146] In addition, the present invention also discovered an interesting phenomenon that when the value of M is low, the zero forcing (ZF) scheme shows better performance improvement, and under the same configuration, as the number of units increases, the scheme proposed by the present invention is better than the ZF scheme. This is because the basic design goal of the ZF scheme is to eliminate interference between users. Therefore, even if the control capability of STAR-RIS is limited, it can still achieve significant performance improvement. On the contrary, when the control capability of STAR-RIS is limited, that is, when M is small, the scheme proposed by the present invention faces challenges in eliminating interference between users. As M increases, STAR-RIS's ability to adjust the input signal is enhanced, which can not only effectively eliminate interference between users, but also improve the quality of signals received by different users, thereby further improving the communication rate. In addition, under the same conditions, by comparing with the simulation results of the SDR scheme, the significant advantages of the algorithm proposed by the present invention are demonstrated, and the effectiveness of the algorithm proposed by the present invention is verified. Please note that compared with other benchmark schemes, the traditional RIS-assisted system shows the smallest performance improvement, even compared with the system with more stringent system conditions (i.e., N t =5) is worse than the proposed solution of the present invention. This shows the remarkable capability of STAR-RIS in effectively realizing full spatial signal modulation.

[0147] Example 2

[0148] Reference Figure 3 In order to verify the effectiveness of the proposed PLS scheme in suppressing information leakage in the full spatial coverage area of ​​the STAR-RIS assisted wireless communication system, the present invention first generates 500 random eavesdropping channels and calculates the corresponding eavesdropping rates with and without security policies. This is done based on optimized active and passive beamforming solutions. Figure 3(a) and (b) represent the eavesdropping rates in the R region and the T region, respectively. As can be seen from the figure, compared with the result without considering the security strategy, the eavesdropping rate of the PLS scheme proposed in the present invention under different CSIs of the eavesdropping user Eve is significantly suppressed, which shows that the robust full-space PLS scheme proposed in the present invention can effectively suppress the eavesdropping of the eavesdropping user Eve on the communication between the base station and the secure user Bob. In addition, the present invention can find that these eavesdropping rates are still significantly increased in the channels of some eavesdropping user Eve, which is mainly attributed to the high coupling of these channels with the communication channel of the secure user Bob. These simulation results also show that the reflection area is more susceptible to eavesdropping than the transmission area, which can be proved by the eavesdropping rate value.

[0149] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0150] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A full-space PLS transmission method for a STAR-RIS assisted wireless communication system, characterized in that: The reflection coefficient and transmission coefficient of STAR-RIS in the STAR-RIS assisted wireless communication system and the precoding vectors allocated by the base station to the security users and public users are obtained by solving a non-convex optimization problem; Among them, the non-convex optimization problem takes the weighted sum of the security rate of the security user and the achievable rate of the public user as the optimization goal, and the base station power as the constraint condition; the security rate of the security user is obtained by performing a large-system analysis on the average security rate of the security user.

2. The full-space PLS transmission method for a STAR-RIS assisted wireless communication system according to claim 1, characterized in that: The non-convex optimization problem is as follows: Where ω1,ω2∈(0,1] represent the weighting factors of secure users and public users respectively; P tmax Indicates the maximum transmit power at the base station; λ represents the number of quantization bits; and Respectively represent the precoding vectors allocated by the base station to the security user and the public user; U r and U t denote the reflection coefficient and transmission coefficient of STAR-RIS respectively; and denote the magnitudes of the STAR-RIS reflection and transmission coefficients, respectively, satisfying are the phases of the STAR-RIS reflection coefficient and transmission coefficient respectively; R c represents the achievable communication rate for public users; Indicates the security rate of the security user.

3. The full-space PLS transmission method for a STAR-RIS assisted wireless communication system according to claim 2, characterized in that: Safe rate for safe users for: Among them, H BR Indicates the channel between the base station and STAR-RIS; represents the noise power at the eavesdropping user; R b Indicates the achievable communication rate of secure users; l re represents the large-scale path loss experienced by the signal from STAR-RIS to the eavesdropping user; P1 represents the probability that the eavesdropping user is located in the R region of STAR-RIS; P0 represents the probability that the eavesdropping user is located in the T region of STAR-RIS, P0=1-P1.

4. The full-space PLS transmission method for a STAR-RIS assisted wireless communication system according to claim 3, characterized in that: The achievable communication rate R of secure users b and the achievable communication rate R of public users c They are as follows: Among them, h rb and h rc The channels between STAR-RIS and security users and public users are shown respectively. and represent the noise power at the safety user and the public user respectively.

5. The full-space PLS transmission method for a STAR-RIS assisted wireless communication system according to claim 4, characterized in that: The method for solving the non-convex optimization problem is: Decomposing the non-convex optimization problem into an active beamforming subproblem and a passive beamforming subproblem, and solving the active beamforming subproblem and the passive beamforming subproblem respectively; The active beamforming subproblem is expressed as: The passive beamforming subproblem is expressed as:

6. The full-space PLS transmission method for a STAR-RIS assisted wireless communication system according to claim 5, characterized in that: The method to solve the active beamforming subproblem is: The minimum mean square error based iterative algorithm is used to solve the active beamforming subproblem, where the active beamforming subproblem is expressed as the following convex optimization problem in the (t+1)th iteration: and is the auxiliary variable in the tth iteration; The convex optimization problem is solved using the Lagrange multiplier method to obtain the optimal solution.

7. The full-space PLS transmission method for a STAR-RIS assisted wireless communication system according to claim 5, characterized in that: The method for solving the passive beamforming subproblem is: using a cross-entropy optimized iterative algorithm to solve the passive beamforming subproblem.

8. The full-space PLS transmission method for a STAR-RIS assisted wireless communication system according to claim 7, characterized in that: The joint sampling distribution of the cross entropy optimization framework is constructed as: in, represents the sampled solution, represents the skew parameter of the joint sampling distribution; the matrix P pq Represents φ p Select Collection The qth item in the The probability of {·} An indicator function representing an event, represents the pth entry of vector a; and denote the standard deviation and mean vector, σ m and μ m is the parameter that generates the amplitude of the mth unit of STAR-RIS; and denote the magnitudes of the STAR-RIS reflection and transmission coefficients, respectively, satisfying represents a discrete phase shift variable; denote the phases of the STAR-RIS reflection coefficient and transmission coefficient, respectively.

9. The full-space PLS transmission method for a STAR-RIS assisted wireless communication system according to claim 8, characterized in that: Based on the joint sampling distribution, the cross entropy minimization optimization problem is: The Lagrange multiplier method is used to solve the above cross entropy minimization optimization problem and obtain the optimal solution.

10. A STAR-RIS assisted wireless communication system, characterized in that: The reflection coefficient and transmission coefficient of STAR-RIS in the STAR-RIS assisted wireless communication system and the precoding vectors allocated by the base station to the security user and the public user are obtained by solving a non-convex optimization problem; Among them, the non-convex optimization problem takes the weighted sum of the security rate of the security user and the achievable rate of the public user as the optimization goal, and the base station power as the constraint condition; the security rate of the security user is obtained by performing a large-system analysis on the average security rate of the security user.

Images